GroupGuard: A Framework for Modeling and Defending Collusive Attacks in Multi-Agent Systems

2026-03-14

AI Summary

This work addresses the emerging threat of collusion attacks in multi-agent systems, where large language model (LLM) agents exploit sociological strategies to coordinate malicious behavior, increasing attack success rates by up to 15% and severely compromising system security and collaborative reliability. To counter this, the paper presents the first formal model of multi-agent collusion attacks and introduces GroupGuard, a novel real-time defense framework that requires no retraining. GroupGuard integrates dynamic graph-structure monitoring, proactive honeypot decoys, and structural pruning mechanisms, leveraging social network analysis to enable efficient detection and isolation of colluding agents. Extensive experiments across five datasets and four network topologies demonstrate that GroupGuard achieves an 88% detection accuracy and effectively restores system-wide collaboration performance.

Technology Category

• Multiagent Systems: Coordination and Collaboration
• Machine Learning: Large Multimodal Models (LMMs)
• Game Theory and Economic Paradigms: Coordination and Collaboration

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

While large language model-based agents demonstrate great potential in collaborative tasks, their interactivity also introduces security vulnerabilities. In this paper, we propose and model group collusive attacks, a highly destructive threat in which multiple agents coordinate via sociological strategies to mislead the system. To address this challenge, we introduce GroupGuard, a training-free defense framework that employs a multi-layered defense strategy, including continuous graph-based monitoring, active honeypot inducement, and structural pruning, to identify and isolate collusive agents. Experimental results across five datasets and four topologies demonstrate that group collusive attacks increase the attack success rate by up to 15\% compared to individual attacks. GroupGuard consistently achieves high detection accuracy (up to 88\%) and effectively restores collaborative performance, providing a robust solution for securing multi-agent systems.

Problem

• collusive attacks
• multi-agent systems
• security vulnerabilities
• group collusion
• LLM-based agents

Innovation

• group collusive attacks
• multi-agent security
• training-free defense
• graph-based monitoring
• honeypot inducement